THE JOURNAL OF CHEMICAL PHYSICS 124, 034311 共2006兲

Comparing reactions of H and Cl with C–H stretch-excited CHD3 Jon P. Camden, Hans A. Bechtel, Davida J. Ankeny Brown, and Richard N. Zarea兲 Department of Chemistry, Stanford University, Stanford, California 94305-5080

共Received 6 September 2005; accepted 22 November 2005; published online 20 January 2006兲 We report the methyl radical product state distributions for the reactions of H and Cl with CHD3共␯1 = 1 , 2兲 at collision energies of 1.53 and 0.18 eV, respectively. Both reactions demonstrate mode selectivity. The resulting state distributions from the H + CHD3共␯1 = 1 , 2兲 reactions are well described by a spectator model. The reactions Cl+ CHD3共␯1 = 1 , 2兲 exhibit similar behavior, but in some aspects the spectator model breaks down. We attribute this breakdown to enhanced intramolecular vibrational redistribution in the Cl+ CHD3共␯1 = 1 , 2兲 reactions compared to the H + CHD3共␯1 = 1 , 2兲 reactions, caused by the interaction of the slower Cl atom with the vibrationally excited CHD3, which is promoted either by its longer collision duration, its stronger coupling, or both. © 2006 American Institute of Physics. 关DOI: 10.1063/1.2155434兴 I. INTRODUCTION

The outcome of a chemical reaction can be controlled, in certain circumstances, by using a laser to excite specific molecular vibrations.1–3 In a series of experiments Crim and co-workers4–10 and Zare and co-workers11–13 demonstrated the preferential cleavage 共bond selectivity兲 of the O–H or O–D bond in the reactions of HOD with H, O, and Cl atoms by exciting either the O–H or O–D stretch, respectively. Crim and co-workers also observed mode-specific behavior in the reactions of H with H2O prepared in the nearly isoenergetic 兩04典− or 兩13典− states, where 兩ab典 is a shorthand for the number of quanta 共a and b兲 in each O–H bond. Excitation of the 兩04典− state produced mainly OH共␯ = 0兲 whereas excitation of the 兩13典− state produced mainly OH共␯ = 1兲. A simple spectator model, in which the vibrational energy in the unreactive bond does not participate in the reaction, was proposed to account for the observed bond-selective and mode-selective behaviors. This spectator model was also found to be qualitatively correct for the reactions of chlorine with vibrationally excited methane.14–21 Trajectory calculations of Schatz et al.22 were qualitatively able to reproduce the bond-selective behavior and demonstrated that the reaction H + HOD共␯O–D = 7兲 → H2 + OD is enhanced by several orders of magnitude over that of the ground state, even though the excitation is localized in the unreactive bond. These results suggest a breakdown of the spectator model: vibrational motion localized in the OD oscillator is able to enhance the reactivity of H-atom abstraction in the H + HOD reaction. The failure of the spectator model is intimately connected with intramolecular vibrational redistribution23 共IVR兲, i.e., the way energy flows between the different internal modes of a molecule. Even if a vibrational eigenstate can be prepared in the reactant valley, the initially prepared vibrational motion might be partitioned into other modes as the system progresses over the barrier and onto the product valley. The details of this energy flow a兲

Author to whom correspondence should be addressed. Electronic mail: [email protected]

0021-9606/2006/124共3兲/034311/7/$23.00

are determined by the amount of time available for the interaction of the incoming atom with the excited reagent and the coupling between the modes of the reactive complex. This description assumes, of course, that motion on only one potential-energy surface suffices to describe the reaction dynamics, i.e., nonadiabatic behavior is assumed to be negligible. Our laboratory in collaboration with Schatz and coworkers has recently performed an extensive study of the H + CD4共␯ = 0兲 reaction.24–26 In that work we examined the CD3 product state and angular distributions as a function of collision energy, comparing them to predictions from various full-dimensional potential-energy surfaces. We have also measured the vibrational enhancement factor and CH3 state distributions for the H + CH4共␯3兲 reaction.27 Attention is drawn to a recent review by Murray and Orr-Ewing28 for a compilation of work on the Cl+ CH4 reaction before 2004. Since that time, however, several new studies have been completed by Liu and co-workers,29–31 Crim and co-workers,32 Orr-Ewing and co-workers,33 and Zare and co-workers.18,34,35 In this work we investigate the CD3 / CHD2 product state distributions from the H + CHD3共␯1 = 1 , 2兲 and Cl+ CHD3共␯1 = 1 , 2兲 reactions, where ␯1 is the C–H stretching vibration. We observe that the H-atom reactions are closer to the pure spectator limit, whereas the Cl-atom reactions are more bond selective. Our results illustrate that the identity of the attacking atom can dramatically influence the bond and mode selectivities observed for the same initially prepared methane vibration. The observed differences suggest a redistribution of the initially prepared vibration during the course of the reactive encounter. A. Infrared spectroscopy of CHD3„␯1 = 1…

Normal modes are usually used to describe the vibrational motions in a polyatomic molecule.36 Because the normal modes of a molecule constitute a complete basis set, any arbitrary motion can be described by the linear superposition of these modes. The CHD3 molecule belongs to the C3␯ point

124, 034311-1

© 2006 American Institute of Physics

Downloaded 16 May 2007 to 165.124.164.50. Redistribution subject to AIP license or copyright, see http://jcp.aip.org/jcp/copyright.jsp

034311-2

J. Chem. Phys. 124, 034311 共2006兲

Camden et al. TABLE I. Normal-mode frequencies in cm−1 of CHD3 , CHD2 共Ref. 61兲 and CD3 共Ref. 61兲. CHD2

CHD3 2993 2142 1003 2263 1291 1036

共 ␯ 1兲 共 ␯ 2兲 共 ␯ 3兲 共 ␯ 4兲 共 ␯ 5兲 共 ␯ 6兲

C–H symmetric stretch C–D symmetric stretch umbrella C–D antisymmetric stretch rock deformation

3114 共␯1兲 C–H stretch 2158 共␯1兲 symmetric stretching 2187 共␯2兲 C–D asymmetric stretch 458 共␯2兲 umbrella bending 1006 共␯3兲 scissors 2381 共␯3兲 antisymmetric stretching 431 共␯4兲 out of plane 1026 共␯4兲 deformation 2358 共␯5兲 C–D asymmetric stretch 1248 共␯6兲 C–H bend

group and has six normal modes of vibration, which are listed in Table I. Upon isotopic substitution of CH4, the totally symmetric ␯1 mode transforms to the totally symmetric ␯1 mode in CHD3 and becomes infrared active. The 2␯1共A1兲 mode of CHD3 is also accessible by one-photon IR absorption. For electric-dipole-allowed transitions between nondegenerate levels in molecules of C3␯ symmetry, the ⌬K = 0 selection rule leads to a parallel band, i.e., the transition dipole moment lies along the symmetry axis, and the molecular transition displays simple P , Q, and R branches. The K sublevels are beyond the resolution of the IR laser used in this experiment. Some normal modes are isolated in a particular bond or region of a molecule; e.g., the ␯1 mode of CHD3 corresponds mainly to stretching of the C–H bond due to the large mass difference between the H and D atoms. More generally, the X–H stretching vibrations and their overtones have proven to be particularly good examples of localized vibrations. A desire to model this behavior and the recognition that one is not restricted to using the normal-mode basis set has led theorists to develop the local-mode description of vibrational modes, in which each bond is treated as an independent anharmonic oscillator.37–43 In general, localization of the vibration occurs when the interbond coupling is weak and the bond anharmonicity is large. The infrared spectroscopy of the C–H chromophore has been the subject of detailed investigations44–47 and is known to be well described by the local-mode picture.48 In particular, the CHD3共␯1 = 1 , 2兲 vibration is localized in the C–H oscillator and in the local-mode basis set we denote CHD3共␯1 = 1兲 as 兩1000典, whereas CHD3共␯1 = 2兲 is

FIG. 1. Energetics for the H + CHD3 reaction.

CD3

given by 兩2000典. Our previous work has illustrated that the local-mode basis set is particularly useful in understanding the reactions of stretch-excited methane with both Cl 共Refs. 15 and 49兲 and H.27 B. Reaction energetics

The H + CH4 → CH3 + H2 reaction is nearly thermoneutral 关⌬H共0 K兲 = −9 ⫻ 10−4 eV兴 共Ref. 50兲 and has a large classical barrier to reaction 关0.64 eV calculated at the CCSD共T兲 level with complete basis set extrapolation using CCSD共T兲/ cc-pVTZ geometries兴.24 The Cl+ CH4 reaction, on the other hand, is slightly endoergic51 共⌬H = 0.07 eV兲 and has an estimated activation barrier52 of 0.34 eV. The vibrationally adiabatic ground-state barrier, which is a better predictor of threshold energies neglecting tunneling, of 0.17 eV.52,53 To calculate the energetics for the H + CHD3 and Cl+ CHD3 reactions, we use the harmonic approximation, noting selected product state energies. The normal-mode frequencies of CD3 and CHD2 are given in Table I. The results of these calculations are shown in Fig. 1 for the H + CHD3 reaction and in Fig. 2 for the Cl+ CHD3 reaction. We note that the H-atom and Cl-atom reactions have markedly different collision energies 共Ecoll兲, 1.52 and 0.18 eV, respectively, and therefore these two reactions have different energetically allowed product state channels. This situation arises because of the mass combinations of the photolytic precursors for the two reactions as well as the mass combinations of the two reactions.54

FIG. 2. Energetics for the Cl+ CHD3 reaction.

Downloaded 16 May 2007 to 165.124.164.50. Redistribution subject to AIP license or copyright, see http://jcp.aip.org/jcp/copyright.jsp

034311-3

Reactions of H and Cl with CHD3

J. Chem. Phys. 124, 034311 共2006兲

FIG. 3. REMPI spectra of the CD3 共left panel兲 and CHD2 共right panel兲 products from the H + CHD3共␯ = 0兲 共top traces兲, H + CHD3共␯1 = 1兲 共middle traces兲, and H + CHD3共␯1 = 2兲 共bottom traces兲 reactions at a collision energy of 1.53 eV. The CD3 and CHD2 spectra are recorded simultaneously for each reaction.

II. EXPERIMENT

The experimental apparatus has been described in detail elsewhere;55 therefore, only the most salient features are described here. Hydrogen bromide 共Matheson, 99.999%兲 or molecular chlorine 共Matheson, 99.999%兲, methane-d3 共Cambridge Isotope Laboratories, 98%兲, and helium 共Liquid Carbonic, 99.995%兲 are mixed in a glass bulb and delivered to a pulsed supersonic nozzle 共General Valve, Series 9, 0.6 mm orifice, backing pressure ⬃700 torr兲. The resulting molecular beam enters the extraction region of a Wiley-McLaren timeof-flight spectrometer where it is intersected by three laser beams that prepare the reagent quantum state, initiate the reaction, and state-selectively probe the products. The CHD3 symmetric stretching fundamental or overtone is prepared by direct infrared absorption around 3.3 or 1.7 ␮m respectively. Fast H atoms are generated by the 230 nm photolysis of HBr,56 and fast Cl atoms are generated from the 355 nm photolysis of Cl257 The photolysis of Cl2 at 355 nm produces monoenergetic Cl atoms in their ground electronic state. The photolysis of HBr at 230 nm produces mainly fast H atoms, i.e., those coincident with ground-state Br; however, a small fraction 共⬃15% 兲 comes with spin-orbit excited Br* and is referred to as the slow channel. This channel might be a cause for some concern; however, we have made a study of the collision energy dependence of both the vibrational enhancement and the methyl radical state distribution for the related H + CH4共␯3 = 1 , 2兲 → CH3 + H2 reaction, and we found no change over the 1.5–2.2 eV energy range.27 Therefore, we feel that the small contribution of the slow channel to these current experiments does not significantly affect our conclusions. After a time delay of 20–30 ns for the H-atom reaction and 70–100 ns for the Cl-atom reaction, the nascent CD3 and CHD2 reaction products are stateselectively ionized using a 2 + 1 resonance-enhanced multi-

photon ionization 共REMPI兲 scheme via the 3pz 2A2⬙ ← X 2A2⬙ transition58 for CD3 and via the 3p 2B1 ← X 2B1 transition59 for CHD2. In order to ensure that no bias exists in the measurements from faster moving products flying out of the probe volume before the slower moving ones, all measurements were made at a time delay for which the CD3 / CHD2 product signal was still a linear function of the time delay. The product ions separate according to their mass and are detected by microchannel plates. In the current experiments, both the m / z = 17 and 18 mass peaks were recorded simultaneously as a function of the REMPI laser wavelength. Large extraction fields 共800 V / cm兲 are used while scanning the REMPI spectra in order to collect all ions of a given mass that are formed in the focal volume of the probing laser. To distinguish between the CD3 / CHD2 products from the reaction of mainly ground-state methane in the molecular beam and the CD3 / CHD2 products from the reaction with vibrationally excited methane, the IR light is modulated on and off on a shot-by-shot basis. Subtraction of the signals that result when the IR laser is on 共Son兲 and off 共Soff兲 gives the difference signal 共Son − Soff兲, which is a measure of the enhancement from vibrational excitation of the CHD3 reagent. Excitation of CHD3共␯1 = 1兲 requires light around 3.3 ␮m, whereas CHD3共␯1 = 2兲 requires light around 1.7 ␮m. Tunable infrared light around 1.7 ␮m is generated by mixing the visible output of a Nd3+: YAG 共yttrium aluminum garnet兲 共Continuum PL9020兲 pumped dye laser 共Continuum, ND6000; Exciton, DCM兲 with the 1.064 ␮m YAG fundamental in a beta barium borate 共BBO兲 crystal. The 1.7 ␮m light is then parametrically amplified in a LiNbO3 crystal which is pumped by 1.064 ␮m radiation. Using this scheme, we obtained ⬃20 mJ after the difference frequency stage and ⬃55 mJ after amplification. The same scheme was used to obtain 3.3 ␮m light; however, instead of using the

Downloaded 16 May 2007 to 165.124.164.50. Redistribution subject to AIP license or copyright, see http://jcp.aip.org/jcp/copyright.jsp

034311-4

Camden et al.

amplified 1.7 ␮m light from the LiNbO3 optical parametric amplification stage, the 3.3 ␮m beam 共⬃12 mJ兲 was used. The 230 nm light 共3–5 mJ兲 was generated by frequency tripling in two BBO crystals the output of a Nd3+: YAG- 共Continuum PL8020兲 pumped dye laser 共Spectra Physics, PDL3; Exciton, LDS 698兲. The ⬃330 nm REMPI probe light 共1.5 mJ兲 was generated by frequency doubling in a BBO crystal the output of a Nd3+: YAG- 共Spectra Physics DCR-2A兲 pumped dye laser 共Lambda Physik, FL2002; Exciton, DCM/ LDS698 mixture兲. III. RESULTS A. Reactions of H with CHD3„␯1 = 0 , 1 , 2…

Figure 3 displays the REMPI spectra obtained for the CD3 / CHD2 products from the reactions of H with CHD3共␯1 = 0 , 1 , 2兲 at Ecoll = 1.53 eV. The signal from the ground-state reaction, obtained when the IR laser is off, originates from the reactions with vibrationally unexcited CHD3. The difference signal, Son − Soff, is shown for when the IR laser pumps the Q branch of the CHD3共␯1 = 1兲 and CHD3共␯1 = 2兲 transitions. The CD3 and CHD2 products are monitored simultaneously as the REMPI laser is scanned by detecting both the m / z = 17 and 18 fragments. All spectra are obtained under similar experimental conditions; thus, although quantitative determinations of the state distribution are complicated owing to a modest signal-to-noise ratio and the effect of power broadening,60 we believe that we can make some meaningful comparisons between the spectra. The ground-state reaction 关Fig. 3共a兲兴 shows no clear preference for the H- or D-abstraction products. We note, however, that isotopic substitution of the CD3 to CHD2 is expected to decrease the sensitivity of the CHD2 REMPI transitions owing to the larger amount of predissociation59 of the intermediate electronic state. Thus, the direct comparison of the exact branching ratio between H / D abstraction is difficult. The qualitative behavior, however, is clear especially in light of the spectra that result from the reactions of H / Cl with vibrationally excited CHD3, vide infra. Both reaction channels produce methyl fragments in their ground state or with lowfrequency bending excitation, i.e., umbrella bending CD3共211兲 and out-of-plane large amplitude 共OPLA兲 for CHD2共411兲. Small features are also observed that correspond to CD3 fragments with several quanta of bending 共222 and 213兲 and to CHD2 fragments with C–D bending excitation 共311兲. The major ground-state reaction channels can be summarized as H + CHD3共␯1 = 0兲 → CD3共␯ = 0, ␯2 = 1,2兲 + H2 → CHD2共␯ = 0, ␯4 = 1兲 + HD. Upon vibrational excitation of the C–H stretching chromophore dramatic changes are observed. Several new features appear in the spectra of the CD3 and CHD2 product channels. Higher overtones of the CD3 umbrella bending motion are observed as hot bands 共213 , 224, and 235兲 whose fraction increases from H + CHD3共␯1 = 1兲 to H + CHD3共␯1 = 2兲. Even more striking differences appear in the CHD2 spectra. A large depletion is observed on the 000 and 411 bands;

J. Chem. Phys. 124, 034311 共2006兲

therefore, the cross section for forming ground-state and OPLA CHD2 products is smaller for the vibrationally excited CHD3 molecules. We also observe the formation of C–H stretch-excited CHD2 in the appearance of the 111 band. The major reaction channels for the fundamental excited reaction can be summarized as H + CHD3共␯1 = 1兲 → CD3共␯ = 0, ␯2 = 1,2,3,4兲 + H2 → CHD2共␯1 = 1兲 + HD → ” CHD2共␯ = 0, ␯4 = 1兲 + HD. This trend continues with excitation of the first C–H stretching overtone of CHD3. In the CHD2 spectrum, the 000 depletion signal remains, the 111 band is no longer present, and a new band, which we attribute to 122, appears. This assignment is made on the basis of the calculated vibrational frequencies of the ground and excited electronic states of CHD2 and the known selection rules from Brum et al..59 The depletion for the CHD3共␯1 = 2兲 spectrum is likely smaller because it is harder to saturate the IR pumping step. One important consideration in the above spectra is what fraction of the methane molecules is pumped to the excited state with the IR laser. From the observed depletion signal on the CHD2 000 band we estimate the fraction N关CHD3共␯1 = 1兲兴 / N关CHD3共␯ = 0兲兴 to be greater than 0.2 and N关CHD3共␯1 = 2兲兴 / N关CHD3共␯ = 0兲兴 to be greater than 0.1, where N is the number of molecules. We note that this estimate does not rely on any assumptions about the spectator model but rather is derived simply from the ratio of the signals with the IR pump laser on and off. The major reaction channels for the overtone excited reaction are summarized as H + CHD3共␯1 = 2兲 → CD3共␯ = 0, ␯2 = 1,2,3,4,5兲 + H2 → CHD2共␯1 = 2兲 + HD → ” CHD2共␯ = 0, ␯4 = 1兲 + HD.

B. Reactions of Cl with CHD3„␯1 = 0 , 1 , 2…

State distributions and angular distributions for the HCl fragment have previously been reported by Simpson et al.14 for the Cl+ CHD3共␯1 = 1兲 reaction and we do not focus on them in this work. Figure 4 displays the CD3 and CHD2 REMPI spectra obtained for the Cl+ CHD3共␯1 = 0 , 1 , 2兲 reactions at Ecoll = 0.18 eV with similar conditions as Fig. 3. Several differences are noted immediately. The ground-state reaction, Cl+ CHD3共␯ = 0兲, produces methyl fragments in their ground vibrational state only, due to energetic constraints, and the ratio of the CD3 products to the CHD2 products is larger. The difficulties in determining the quantitative H / D abstraction ratio presented in Sec. III A do not affect this conclusion, as the REMPI spectra were obtained under similar experimental conditions. The fact that the energy available to the Cl-atom reaction is much closer to threshold suggests that the difference in zero-point energies of the C–H and C–D bonds could explain the larger CD3 / CHD2 ratio in the Cl-atom reaction when compared to the H-atom reaction.

Downloaded 16 May 2007 to 165.124.164.50. Redistribution subject to AIP license or copyright, see http://jcp.aip.org/jcp/copyright.jsp

034311-5

J. Chem. Phys. 124, 034311 共2006兲

Reactions of H and Cl with CHD3

FIG. 4. REMPI spectra of the CD3 共left panel兲 and CHD2 共right panel兲 products from the Cl+ CHD3共␯ = 0兲 共top traces兲, Cl+ CHD3共␯1 = 1兲 共middle traces兲, and Cl+ CHD3共␯1 = 2兲 共bottom traces兲 reactions at a collision energy of 0.18 eV. The CD3 and CHD2 spectra are recorded simultaneously for each reaction.

The main reaction channels are

IV. DISCUSSION A. Spectator model

Cl + CHD3共␯ = 0兲 → CD3共␯ = 0兲 + HCl → CHD2共␯ = 0兲 + DCl. Upon C–H stretch excitation of CHD3 the CD3 products are formed primarily in their ground state, in contrast to the H-atom reaction. A small but increasing fraction of the CD3 products is bend excited as the number of C–H stretching quanta of the methane reagent increases. The reaction of CHD3共␯1 = 1兲 with Cl exhibits a strong preference for the H-abstraction channel, but a very small amount of the channel leading to CHD2共␯1 = 1兲 is observed. Lastly, the reaction of Cl with CHD3共␯1 = 2兲 leads to similar behavior, except that no stretch-excited CHD2 is observed, although this may simply be caused by a lack of sensitivity. Summarizing the reaction channels of Cl with CHD3共␯1兲, Cl + CHD3共␯1 = 1兲 → CD3共␯ = 0, ␯2 = 1,2,3兲 + HCl → 共minor兲CHD2共␯1 = 1兲 + DCl, Cl + CHD3共␯1 = 2兲 → CD3共␯ = 0, ␯2 = 1,2,3兲 + HCl. The most striking differences between the H- and Clatom reactions occur upon C–H stretch excitation of CHD3, as can be seen by comparing Figs. 3 and 4. The H-atom reaction with CHD3共␯1兲 leads to a depletion of the groundstate CHD2 products, production of C–H stretch-excited CHD2, and excitation of the CD3 bending modes. The same initially prepared vibration for the Cl-atom reaction shows no depletion signal, an extremely small amount of C–H stretch-excited CHD2, and less preference for CD3 bending mode excitation.

We begin by briefly reviewing the spectator model and giving a clear description of the assumptions we use in the following discussion. For a polyatomic molecule such as methane we define the pure spectator limit as one in which every bond acts as a local uncoupled oscillator. In this crude picture we neglect bending motions and assume that every C–H bond is independent of all others. Further, the initial vibration is given by the local-mode description. The CHD3共␯1 = 1 , 2兲 vibration is localized in the C–H oscillator. In the local-mode basis set we denote CHD3共␯1 = 1兲 as 兩1000典 and CHD3共␯1 = 2兲 as 兩2000典. In this simple picture, the H / Cl atom has a choice when it approaches the vibrating methane. Reaction with the C–H oscillator will leave the CD3 fragment in its ground vibrational state, whereas reaction with a C–D bond will leave CHD2 fragments with one quantum of C–H stretching. Thus, there is no mechanism for the formation of ground-state CHD2 products. This model applies equally to the reaction of 兩2000典 excited CHD3, except in this case reaction with a C–D oscillator will lead to CHD2 products with two quanta of CH stretching. An interesting prediction of this model is that the reaction cross section for forming ground-state CHD2 fragments should actually be smaller in the vibrationally excited reactions, a point that has not been addressed until this work. Of course, we might also ask if the initially localized excitation can facilitate cleavage of the unexcited C–D bonds, as was suggested by the trajectory calculations of Schatz et al.22 Table II displays the state-selected ratios of ␴␯1 / ␴gs that result from predictions of the isolated bond model. If the ratio is greater than 共⬎兲, equal to 共=兲, or less than 共⬍兲 1, the vibrationally excited cross section is larger, equal to, or smaller than the ground-state reaction, respectively. For sim-

Downloaded 16 May 2007 to 165.124.164.50. Redistribution subject to AIP license or copyright, see http://jcp.aip.org/jcp/copyright.jsp

034311-6

J. Chem. Phys. 124, 034311 共2006兲

Camden et al.

TABLE II. Observed ratio of the vibrationally excited to ground-state reaction cross section. If the ratio is greater than 共⬎兲, equal to 共=兲, or less than 共⬍兲 1, the vibrationally excited cross section is larger, equal to, or smaller than the ground-state reaction, respectively. The expected value of this ratio obtained from the spectator model is given for comparison. It is not possible to put the experiment on an absolute scale, so that the prediction of 0 is in accord with the depletion of the product, which is indicated by ⬍. The H-atom reactions are seen to be in good agreement with the model, whereas the Cl-atom reactions show some clear disagreements, which are indicated by the asterisk 共*兲. CHD2共␯ = 0兲

H + CHD3兩1000典 Cl+ CHD3兩1000典 H + CHD3兩2000典 Cl+ CHD3兩2000典

CHD2共␯1 = 1兲

CHD2共␯1 = 2兲

CD3共␯ = 0兲

Model

Expt.

Model

Expt.

Model

Expt.

Model

Expt.

0 0* 0 0*

⬍ =* ⬍ =*

⬎ ⬎* 0 0

⬎ ⬃0* 0 0

0 0 ⬎ ⬎

0 0 ⬎ 0*

⬎ ⬎ ⬎ ⬎*

⬎ ⬎ ⬎ ⬎

plicity, we exclude the umbrella bending products in this treatment. A zero indicates that the product state is not expected for a given reaction channel. B. Comparing the H- and Cl-atom reactions

We first consider the reactions of H and Cl to form ground-state CHD2. Recall that Soff ⬀ ␴␯=0 and Son ⬀ 共1 − x兲␴␯=0 + x␴IR and their ratio is given by

␴IR 共Son/Soff − 1兲 + 1. = ␴␯=0 x Therefore, if Son / Soff = 1, which results when Son − Soff = 0, then the ratio of the cross sections is also unity and the ground-state and vibrationally excited reactions have the same cross section. In practice, it is hard to determine whether a difference signal of zero arises because 共1兲 the ratio of the cross sections is actually unity or 共2兲 the small positive or negative signal is below the experimental sensitivity. However, this ambiguity is removed in the current experiments because of the clear depletion observed for the H + CHD3共␯1 = 1 , 2兲 → CHD2共␯ = 0兲 + HD reactions. The IR pumping scheme was not changed between the two reactions; therefore it is clear that the cross section for forming ground-state CHD2 from the reactions Cl+ CHD3共␯1 = 1 , 2兲 is comparable to that of the Cl+ CHD3共␯ = 0兲 reaction. A positive, zero, or negative signal provides a direct measurement of the ratio of the ground-state and excitedstate cross sections. These measurements are tabulated along with the values expected for the spectator model in Table II. Depletion of a product, denoted by ⬍, is considered to be in rough agreement with the model prediction of 0. It is clear that the H-atom reaction is in better agreement with the model than the Cl-atom reaction, which shows marked disagreements in several cases. For example, the cross sections for formation of ground-state CHD2 from the reaction of C–H stretch-excited and ground-state CHD3 are approximately equal, contrary to the predictions of the spectator model. Therefore, when a D atom is abstracted from C–H stretch-excited CHD3, the vibration must flow from the C–H bond into the translation of the escaping product fragments or the internal excitation of the DCl modes. In either case, the vibration does not remain localized during the reaction, which contradicts the spectator model. This behavior is also

observed for the reaction Cl+ CHD3共␯1 = 2兲 in which the cross section for CHD2共␯ = 0兲 is about the same as that for the reaction Cl+ CHD3共␯ = 0兲. In this case, the breakdown of the spectator model is even more accented as two quanta of C–H stretching must flow from the C–H bond, which leaves the CHD2 product vibrationless. It might be argued that in the normal-mode picture, the v1 vibration is not entirely localized in the C–H bond but has a small amount in the C–D bonds, which might explain the ability of the methane vibration to promote abstraction of the C–D bonds and leave the CHD2 fragment in its ground state in the Cl-atom reaction, but we can rule out such a possibility because of our results for the H-atom reaction. If this were the case, then we might expect to see either 共a兲 a similar enhancement in the H-atom reaction or 共b兲 a small amount of CD3 products formed with stretch excitation. One curious feature that we observe is rather large excitation of the bending modes, up to five quanta in umbrella bending in CD3, which occurs in the H-atom reactions when compared to the Cl-atom reaction. In all cases, the ratio of the umbrella bend-excited CD3 products to the ground-state CD3 is larger for the H-atom reactions, although we point out that the fraction of energy deposited into the bending mode is still small when compared to the total energy available. Our REMPI spectra suggest that the initially prepared vibration is more effectively transferred into the product bending modes in the H-atom reaction than the Cl-atom reaction. Neither the spectator picture nor a simple adiabatic picture of the dynamics is able to rationalize this behavior. Our experiments suggest that excitation of the stretching motion in CHD3 leads nonadiabatically to ground-state methyl fragments in the Cl + CHD3共␯1 = 1 , 2兲 reactions, whereas stretching excitation leads nonadiabatically to bend-excited methyl fragments in H + CHD3共␯1 = 1 , 2兲. The present study clearly indicates that intramolecular vibrational redistribution 共IVR兲 takes place to various extents during the course of the reactions, more for Cl+ CHD3共␯1 = 1 , 2兲 and much less for H + CHD3共␯1 = 1 , 2兲. In our previous study27 of the reactions H / Cl+ CH4共␯3 = 1 , 2兲 we proposed that the difference in interaction time could account for the major observed differences in the Cl- and H-atom reactions. While it is true that the H atom approaches more quickly than the Cl atom and thus leaves less time for the vibration to be localized into the reactive bond, we need to include the

Downloaded 16 May 2007 to 165.124.164.50. Redistribution subject to AIP license or copyright, see http://jcp.aip.org/jcp/copyright.jsp

034311-7

possibility that the Cl atom may couple more strongly than the H atom in the course of the reaction. Both effects may be at work. The exact details are not possible to extract from the present study, and more experiments as well as more theoretical studies are needed to elucidate the nature of the IVR for this chemical reaction. In summary, we have studied the CD3 / CHD2 state distributions that result from the reactions of H and Cl with CHD3共␯1 = 0 , 1 , 2兲. Notable differences exist between the two reactions: particularly, the H-atom reaction appears to be in closer accord with the pure spectator model. A simple explanation based on the greater importance of IVR for Cl + CHD3共␯1兲 is proposed to account for the different behaviors found in these two related reaction systems. Although IVR seems to be more pronounced for the Cl-atom reaction with C–H stretch-excited CHD3, the vibrational redistribution must contribute to the stronger bond selectivity of this reaction. ACKNOWLEDGMENTS

Two of the authors 共J.P.C. and H.A.B.兲 thank the National Science Foundation for graduate fellowships. This work was supported by National Science Foundation Grant No. 0242103. F. F. Crim, J. Phys. Chem. 100, 12725 共1996兲. F. F. Crim, Acc. Chem. Res. 32, 877 共1999兲. R. N. Zare, Science 279, 1875 共1998兲. 4 A. Sinha, M. C. Hsiao, and F. F. Crim, J. Chem. Phys. 92, 6333 共1990兲. 5 A. Sinha, M. C. Hsiao, and F. F. Crim, J. Chem. Phys. 94, 4928 共1991兲. 6 A. Sinha, J. D. Thoemke, and F. F. Crim, J. Chem. Phys. 96, 372 共1992兲. 7 J. M. Pfeiffer, R. B. Metz, J. D. Thoemke, E. Woods, and F. F. Crim, J. Chem. Phys. 104, 4490 共1996兲. 8 J. M. Pfeiffer, E. Woods, R. B. Metz, and F. F. Crim, J. Chem. Phys. 113, 7982 共2000兲. 9 M. C. Hsiao, A. Sinha, and F. F. Crim, J. Phys. Chem. 95, 8263 共1991兲. 10 R. B. Metz, J. D. Thoemke, J. M. Pfeiffer, and F. F. Crim, J. Chem. Phys. 99, 1744 共1993兲. 11 M. J. Bronikowski, W. R. Simpson, B. Girard, and R. N. Zare, J. Chem. Phys. 95, 8647 共1991兲. 12 M. J. Bronikowski, W. R. Simpson, and R. N. Zare, J. Phys. Chem. 97, 2194 共1993兲. 13 M. J. Bronikowski, W. R. Simpson, and R. N. Zare, J. Phys. Chem. 97, 2204 共1993兲. 14 W. R. Simpson, T. P. Rakitzis, S. A. Kandel, A. J. Orr-Ewing, and R. N. Zare, J. Chem. Phys. 103, 7313 共1995兲. 15 Z. H. Kim, H. A. Bechtel, and R. N. Zare, J. Am. Chem. Soc. 123, 12714 共2001兲. 16 H. A. Bechtel, Z. H. Kim, J. P. Camden, and R. N. Zare, J. Chem. Phys. 120, 791 共2004兲. 17 H. A. Bechtel, J. P. Camden, D. J. A. Brown, and R. N. Zare, J. Chem. Phys. 120, 5096 共2004兲. 18 Z. H. Kim, H. A. Bechtel, J. P. Camden, and R. N. Zare, J. Chem. Phys. 122, 084303 共2005兲. 19 S. Yoon, S. Henton, A. N. Zivkovic, and F. F. Crim, J. Chem. Phys. 116, 10744 共2002兲. 20 S. Yoon, R. J. Holiday, and F. F. Crim, J. Chem. Phys. 119, 4755 共2003兲. 21 S. Yoon, R. J. Holiday, E. L. Sibert, and F. F. Crim, J. Chem. Phys. 119, 9568 共2003兲. 22 G. C. Schatz, M. C. Colton, and J. L. Grant, J. Phys. Chem. 88, 2971 共1984兲. 1 2 3

J. Chem. Phys. 124, 034311 共2006兲

Reactions of H and Cl with CHD3

D. J. Nesbitt and R. W. Field, J. Phys. Chem. 100, 12735 共1996兲. J. P. Camden, H. A. Bechtel, D. J. A. Brown, M. R. Martin, R. N. Zare, W. Hu, G. Lendvay, D. Troya, and G. C. Schatz, J. Am. Chem. Soc. 127, 11898 共2005兲. 25 J. P. Camden, W. Hu, H. A. Bechtel, D. J. A. Brown, M. R. Martin, R. N. Zare, G. Lendvay, D. Troya, and G. C. Schatz, J. Phys. Chem. A 共in press兲. 26 W. Hu, G. Lendvay, D. Troya, G. C. Schatz, J. P. Camden, H. A. Bechtel, D. J. A. Brown, M. R. Martin, and R. N. Zare, J. Phys. Chem. A 共in press兲. 27 J. P. Camden, H. A. Bechtel, D. J. A. Brown, and R. N. Zare, J. Chem. Phys. 123, 134301 共2005兲. 28 C. Murray and A. J. Orr-Ewing, Int. Rev. Phys. Chem. 23, 435 共2004兲. 29 J. Zhou, J. J. Lin, B. Zhang, and K. Liu, J. Phys. Chem. A 108, 7832 共2004兲. 30 B. Zhang and K. Liu, J. Chem. Phys. 122, 101102 共2005兲. 31 J. Zhou, B. Zhang, J. J. Lin, and K. Liu, Mol. Phys. 103, 1757 共2005兲. 32 S. Yoon, R. J. Holiday, and F. F. Crim, J. Phys. Chem. B 109, 8388 共2005兲. 33 B. Retail, J. K. Pearce, C. Murray, and A. J. Orr-Ewing, J. Chem. Phys. 122, 101101 共2005兲. 34 H. A. Bechtel, J. P. Camden, D. J. A. Brown, R. M. Martin, R. N. Zare, and K. Vodopyanov, Angew. Chem., Int. Ed. 44, 2382 共2005兲. 35 H. A. Bechtel, Z. H. Kim, J. P. Camden, and R. N. Zare, Mol. Phys. 103, 1837 共2005兲. 36 G. Herzberg, Infrared and Raman Spectra of Polyatomic Molecules 共Van Nostrand, New York, 1945兲. 37 M. S. Child and L. Halonen, Adv. Chem. Phys. 57, 1 共1984兲. 38 M. S. Child, Acc. Chem. Res. 18, 45 共1985兲. 39 B. R. Henry, Acc. Chem. Res. 10, 207 共1977兲. 40 C. Jaffe and P. Brumer, J. Chem. Phys. 73, 5646 共1980兲. 41 M. S. Child and R. T. Lawton, Faraday Discuss. 71, 273 共1981兲. 42 M. M. Law and J. L. Duncan, Mol. Phys. 93 809 共1998兲. 43 P. Jensen, Mol. Phys. 98, 1253 共2000兲. 44 W. E. Blass and T. H. Edwards, J. Mol. Spectrosc. 24, 116 共1967兲. 45 T. A. Wiggins, E. R. Shull, J. M. Bennett, and D. H. Rank, J. Chem. Phys. 21, 1940 共1953兲. 46 M. Lewerenz and M. Quack, J. Chem. Phys. 88, 5408 共1988兲. 47 H.-R. Dubal, T.-K. Ha, M. Lewerenz, and M. Quack, J. Chem. Phys. 91, 6698 共1989兲. 48 L. Halonen and M. S. Child, J. Chem. Phys. 79, 4355 共1983兲. 49 Z. H. Kim, H. A. Bechtel, and R. N. Zare, J. Chem. Phys. 117, 3232 共2002兲. 50 M. W. Chase, Jr., C. A. Davies, J. R. Downey, D. J. Frurip, R. A. McDonald, and A. N. Syverud, J. Phys. Chem. Ref. Data Suppl. 14, 1 共1998兲. 51 R. Atkinson, D. L. Baulch, R. A. Cox, J. R. F. Hampson, J. A. Kerr, and J. Troe, J. Phys. Chem. Ref. Data 21, 1125 共1992兲. 52 Y. Zhao, B. J. Lynch, and D. G. Truhlar, J. Phys. Chem. A 108, 4786 共2004兲. 53 J. C. Corchado, D. G. Truhlar, and J. Espinosa-Garcia, J. Chem. Phys. 112, 9375 共2000兲. 54 N. E. Shafer, A. J. Orr-Ewing, W. R. Simpson, H. Xu, and R. N. Zare, Chem. Phys. Lett. 212, 155 共1993兲. 55 W. R. Simpson, A. J. Orr-Ewing, T. P. Rakitzis, S. A. Kandel, and R. N. Zare, J. Chem. Phys. 103, 7299 共1995兲. 56 P. M. Regan, S. R. Langford, A. J. Orr-Ewing, and M. N. R. Ashfold, J. Chem. Phys. 110, 281 共1999兲. 57 P. C. Samartzis, B. Bakker, T. P. Rakitzis, D. H. Parker, and T. N. Kitsopoulos, J. Chem. Phys. 110, 5201 共1999兲. 58 J. W. Hudgens, T. G. DiGiuseppe, and M. C. Lin, J. Chem. Phys. 79, 571 共1983兲. 59 J. L. Brum, R. D. Johnson III, and J. W. Hudgens, J. Chem. Phys. 98, 3732 共1993兲. 60 J. Zhou, J. J. Lin, W. Shiu, S. C. Pu, and K. Liu, J. Chem. Phys. 119, 2538 共2003兲. 61 A. M. Mebel and S.-H. Lin, Chem. Phys. 215, 329 共1997兲. 23 24

Downloaded 16 May 2007 to 165.124.164.50. Redistribution subject to AIP license or copyright, see http://jcp.aip.org/jcp/copyright.jsp